CSA Social Marketing: News Clusters

Clustering News Topics

“Which topics are alike in terms of their audiences?”

Correlation Matrix

Optimal Number of Clusters

Hierarchical Clustering

K Means

PC1 = Broad/Systemic Issues (left) vs. Everyday/Current Events (right) PC2 = Large scale (bottom) vs. Local scale (top)

Biplot comparing to news gratifications

PC1: uplift; improvement (left); pragmatic (right) PC2: communal, affective (up); cognitive (down)

Biplot comparing to news characteristics

PC1: creative, interpretive, diverse (left); factual, straightforward (right) OC2: light, entertaining (up); serious (down)

Biplot comparing to demographics

Clustering news gratifications

“Which gratifications are alike in terms of their seekers?”

Optimal number of clusters

Hierarchical clustering

K Means

PC1 = Self-enhacement/affirming (left) vs. Civic-altruistic (right) PC2 = Affective/Moral (Top) vs. Cognitive/Pragmatic (Bottom)

Biplot comparing to news topics

Biplot comparing to news characteristics

Biplot comparing to demographics

Clustering news characteristics

“Which characteristics are alike in terms of their admirers?”

Optimal number of clusters

Hierarchical clustering

K Means

Biplot comparing to news topics

Biplot comparing to news gratifications

Biplot comparing to demographics